Resources to be moved
For 1:1 active/standby, N:1 active/standby or cascading configuration, move the following resources to the shared disk.
In the 1:1 active/standby configuration (with or without subsystems), move the following resources to the shared disk. And this task does not need to be done on each subsystem.
For PRIMECLUSTER (for Solaris or Linux version), Oracle Solaris Cluster
The directory for calendar control information
/var/opt/FJSVjmcal/post
The database directory for the Jobscheduler
/var/opt/FJSVJOBSC
The spool directory for Job Execution Control
/var/opt/FJSVMJS/var/spool/mjes
(Delete the symbolic link to "/var/spool/mjes".)
The operation information directory for Job Execution Control
/etc/opt/FJSVMJS/etc/mjes
(Delete the symbolic link to "/etc/mjes".)
The database directory for Master Schedule Management (only if the Master Schedule Management function is enabled)
/var/opt/FJSVstem
For MC/ServiceGuard
The directory for calendar control information
/opt/FHPjmcal/post
The database directory for Jobscheduler
/opt/FHPJOBSCH/db
The spool directory for Job Execution Control
/opt/FHPMJS/var/spool/mjes
(Delete the symbolic link to "/var/spool/mjes".)
The operation information directory for Job Execution Control
/opt/FHPMJS/etc/mjes
(Delete the symbolic link to "/etc/mjes".)
The database directory for Master Schedule Management (only if the Master Schedule Management function is enabled)
/var/opt/FJSVstem
For HACMP
The directory for calendar control information
/opt/FAIXjmcal/post
The database directory for Jobscheduler
/opt/FAIXJOBSC/db
The spool directory for Job Execution Control
/opt/FAIXMJS/var/spool/mjes
(Delete the symbolic link to "/var/spool/mjes".)
The operation information directory for Job Execution Control
/opt/FAIXMJS/etc/mjes
(Delete the symbolic link to "/etc/mjes".)
The database directory for Master Schedule Management (only if the Master Schedule Management function is enabled)
/var/opt/FJSVstem
Relocation procedure
For PRIMECLUSTER (for Solaris or Linux version), Sun Cluster
The explanation in the following example assumes that the mount point of the shared disk is "/disk1". The command line provided here as a sample is for Solaris version.
1) Stop the daemon.
Stop the Systemwalker Operation Manager daemon.
# /opt/systemwalker/bin/poperationmgr -s
For environments that include both Systemwalker Operation Manager and Systemwalker Centric Manager, be sure to specify the "-s" option with the poperationmgr command.
2) Mount the shared disk.
Mount the shared disk.
# mount /disk1 (*1)
The command line for Linux version is as follows. (In this example, the device is "/dev/sdb1".)
# /bin/mount /dev/sdb1 /disk1
3) Move resources from the active node to the shared disk.
Move the resources from the active node to the shared disk, and delete the symbolic links.
# mkdir -p /disk1/FJSVfwseo/JM (*1) # chmod -R 755 /disk1/FJSVfwseo # mkdir -p /disk1/FJSVjmcal # chmod -R 755 /disk1/FJSVjmcal # mv /var/opt/FJSVjmcal/post /disk1/FJSVjmcal/post # mkdir -p /disk1/FJSVJOBSC # chmod -R 755 /disk1/FJSVJOBSC # mv /var/opt/FJSVJOBSC/* /disk1/FJSVJOBSC # rm -r /var/opt/FJSVJOBSC # mkdir -p /disk1/FJSVMJS/var/spool # chmod -R 755 /disk1/FJSVMJS # mv /var/opt/FJSVMJS/var/spool/mjes /disk1/FJSVMJS/var/spool/mjes # rm /var/spool/mjes # mkdir -p /disk1/FJSVMJS/etc # chmod -R 755 /disk1/FJSVMJS # mv /etc/opt/FJSVMJS/etc/mjes /disk1/FJSVMJS/etc/mjes # rm /etc/mjes # mkdir -p /disk1/FJSVstem (*2) # chmod -R 755 /disk1/FJSVstem (*2) # mv /var/opt/FJSVstem/* /disk1/FJSVstem (*2) # rm -r /var/opt/FJSVstem (*2)
For security information, simply a directory is created on the shared disk. This information will be updated by the "2.7 Settings for Automatic Reflection."
This is required only if the Master Schedule Management function is enabled.
4) Delete resources from the standby node.
Delete the resources from the standby node.
# rm -r /var/opt/FJSVjmcal/post # rm -r /var/opt/FJSVJOBSC # rm -r /var/opt/FJSVMJS/var/spool/mjes # rm /var/spool/mjes # rm -r /etc/opt/FJSVMJS/etc/mjes # rm /etc/mjes # rm -r /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
5) Create symbolic links to the relocated resources.
Create symbolic links on all the nodes (active and standby) so that any one of them can use the resources that have been relocated to the shared disk.
For N:1 active/standby configuration, state transition procedures are set up to create symbolic links when failover occurs, so symbolic links from the standby node do not need to be created here.
# ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM # ln -s /disk1/FJSVjmcal/post /var/opt/FJSVjmcal/post # ln -s /disk1/FJSVJOBSC /var/opt/FJSVJOBSC # ln -s /disk1/FJSVMJS/var/spool/mjes /var/spool/mjes # ln -s /disk1/FJSVMJS/etc/mjes /etc/mjes # ln -s /disk1/FJSVstem /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
6) Configure settings for SELinux.
This must be done only in Linux environments where SELinux is enabled. Otherwise, it is not necessary.
Execute the following commands in order to set the SELinux security policies on the relocated resources. They must be executed on each node with the shared disk mounted.
# /usr/sbin/semanage fcontext -a -t var_t '/disk1/FJSVfwseo/JM(/.*)?' # /usr/sbin/semanage fcontext -a -t var_t '/disk1/FJSVjmcal(/.*)?' # /usr/sbin/semanage fcontext -a -t var_t '/disk1/FJSVJOBSC(/.*)?' # /usr/sbin/semanage fcontext -a -t var_t '/disk1/FJSVstem(/.*)?' (*1) # /usr/sbin/semanage fcontext -a -t sw_fjsvmjs_spool_t '/disk1/FJSVMJS/var/spool(/.*)?' # /usr/sbin/semanage fcontext -a -t etc_t '/disk1/FJSVMJS/etc(/.*)?' # /sbin/restorecon -R /disk1
This is required only if the Master Schedule Management function is enabled.
7) Unmount the shared disk.
Unmount the shared disk.
# umount /disk1 (*1)
The command line for Linux version is as follows.
# /bin/umount /disk1
For MC/ServiceGuard
The explanation in the following example assumes that the device of the shared disk is "/dev/vg01/lvol1" and the mount point of the shared disk is "/disk1".
1) Stop the daemon.
Stop the Systemwalker Operation Manager daemon.
# /opt/systemwalker/bin/poperationmgr -s
For environments that include both Systemwalker Operation Manager and Systemwalker Centric Manager, be sure to specify the "-s" option with the poperationmgr command.
2) Move resources from the active node to the shared disk.
Mount the shared disk and move resources from the active node to the shared disk.
Mount the shared disk by using the vgchange command for HP-UX on the active node.
# vgchange -c y /dev/vg01 # vgchange -a e /dev/vg01 # mount /dev/vg01/lvol1 /disk1
Move the resources from the active node to the shared disk, and delete the symbolic links.
# mkdir -p /disk1/FJSVfwseo/JM (*1) # mkdir -p /disk1/FHPjmcal # chmod -R 755 /disk1/FHPjmcal # mv /opt/FHPjmcal/post /disk1/FHPjmcal/post # mkdir -p /disk1/FHPJOBSCH # chmod -R 755 /disk1/FHPJOBSCH # mv /opt/FHPJOBSCH/db/* /disk1/FHPJOBSCH # rm -r /opt/FHPJOBSCH/db # mkdir -p /disk1/FHPMJS/var/spool # chmod -R 755 /disk1/FHPMJS # mv /opt/FHPMJS/var/spool/mjes /disk1/FHPMJS/var/spool/mjes # rm /var/spool/mjes # mkdir -p /disk1/FHPMJS/etc # chmod -R 755 /disk1/FHPMJS # mv /opt/FHPMJS/etc/mjes /disk1/FHPMJS/etc/mjes # rm /etc/mjes # mkdir -p /disk1/FJSVstem (*2) # chmod -R 755 /disk1/FJSVstem (*2) # mv /var/opt/FJSVstem/* /disk1/FJSVstem (*2) # rm -r /var/opt/FJSVstem (*2)
For security information, simply a directory is created on the shared disk. This information will be updated by the "2.7 Settings for Automatic Reflection."
This is required only if the Master Schedule Management function is enabled.
3) Delete resources from the standby node.
Delete the resources from the standby node.
# rm -r /opt/FHPjmcal/post # rm -r /opt/FHPJOBSCH/db # rm -r /opt/FHPMJS/var/spool/mjes # rm /var/spool/mjes # rm -r /opt/FHPMJS/etc/mjes # rm /etc/mjes # rm -r /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
4) Create symbolic links to the relocated resources.
Create symbolic links on both the nodes (active and standby) so that any one of them can use the resources that have been relocated to the shared disk.
For N:1 active/standby configuration, the package control script is set up to create symbolic links when failover occurs, so symbolic links from the standby node do not need to be created here.
Create symbolic links on the active node.
# ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM # ln -s /disk1/FHPjmcal/post /opt/FHPjmcal/post # ln -s /disk1/FHPJOBSCH /opt/FHPJOBSCH/db # ln -s /disk1/FHPMJS/var/spool/mjes /var/spool/mjes # ln -s /disk1/FHPMJS/etc/mjes /etc/mjes # ln -s /disk1/FJSVstem /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
Unmount the shared disk by using the vgchange command for HP-UX on the active node.
# umount /disk1 # vgchange -a n /dev/vg01
Mount the shared disk by using the vgchange command for HP-UX on the standby node.
# vgchange -a e /dev/vg01 # mount /dev/vg01/lvol1 /disk1
Create symbolic links on the standby node.
# ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM # ln -s /disk1/FHPjmcal/post /opt/FHPjmcal/post # ln -s /disk1/FHPJOBSCH /opt/FHPJOBSCH/db # ln -s /disk1/FHPMJS/var/spool/mjes /var/spool/mjes # ln -s /disk1/FHPMJS/etc/mjes /etc/mjes # ln -s /disk1/FJSVstem /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
Unmount the shared disk by using the vgchange command for HP-UX on the standby node.
# umount /disk1 # vgchange -a n /dev/vg01
For HACMP
The explanation in the following example assumes that the volume name of the shared disk is "datavg1" and the mount point of the shared disk is "/disk1".
1) Stop the daemon.
Stop the Systemwalker Operation Manager daemon.
# /opt/systemwalker/bin/poperationmgr -s
For environments that include both Systemwalker Operation Manager and Systemwalker Centric Manager, be sure to specify the "-s" option with the poperationmgr command.
2) Move resources from the active node to the shared disk.
Mount the shared disk and move the resources from the active node to the shared disk.
Activate the volume by using the varyonvg command for AIX on the active node, and then mount the shared disk.
# varyonvg datavg1 # mount /disk1
Move the resources from the active node to the shared disk, and delete the symbolic links.
# mkdir -p /disk1/FJSVfwseo/JM (*1) # chmod -R 755 /disk1/FJSVfwseo # mkdir -p /disk1/FAIXjmcal # chmod -R 755 /disk1/FAIXjmcal # mv /opt/FAIXjmcal/post /disk1/FAIXjmcal/post # mkdir -p /disk1/FAIXJOBSC # chmod -R 755 /disk1/FAIXJOBSC # mv /opt/FAIXJOBSC/db/* /disk1/FAIXJOBSC # rm -r /opt/FAIXJOBSC/db # mkdir -p /disk1/FAIXMJS/var/spool # chmod -R 755 /disk1/FAIXMJS # mv /opt/FAIXMJS/var/spool/mjes /disk1/FAIXMJS/var/spool/mjes # rm /var/spool/mjes # mkdir -p /disk1/FAIXMJS/etc # chmod -R 755 /disk1/FAIXMJS # mv /opt/FAIXMJS/etc/mjes /disk1/FAIXMJS/etc/mjes # rm /etc/mjes # mkdir -p /disk1/FJSVstem (*2) # chmod -R 755 /disk1/FJSVstem (*2) # mv /var/opt/FJSVstem/* /disk1/FJSVstem (*2) # rm -r /var/opt/FJSVstem (*2)
For security information, simply a directory is created on the shared disk. This information will be updated by the "2.7 Settings for Automatic Reflection."
This is required only if the Master Schedule Management function is enabled.
3) Delete resources from the standby node.
Delete the resources from the standby node.
# rm -r /opt/FAIXjmcal/post # rm -r /opt/FAIXJOBSC/db # rm -r /opt/FAIXMJS/var/spool/mjes # rm /var/spool/mjes # rm -r /opt/FAIXMJS/etc/mjes # rm /etc/mjes # rm -r /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
4) Create symbolic links to the relocated resources.
Create symbolic links on both the nodes (active and standby) so that any one of them can use the resources that have been relocated to the shared disk.
Create symbolic links on the active node.
# ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM # ln -s /disk1/FAIXjmcal/post /opt/FAIXjmcal/post # ln -s /disk1/FAIXJOBSC /opt/FAIXJOBSC/db # ln -s /disk1/FAIXMJS/var/spool/mjes /var/spool/mjes # ln -s /disk1/FAIXMJS/etc/mjes /etc/mjes # ln -s /disk1/FJSVstem /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
Unmount the shared disk on the active node and deactivate the volume by using the varyoffvg command for AIX.
# umount /disk1 # varyoffvg datavg1
Activate the volume by using the varyonvg command for AIX on the standby node, and then mount the shared disk.
# varyonvg datavg1 # mount /disk1
Create symbolic links on the standby node.
# ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM # ln -s /disk1/FAIXjmcal/post /opt/FAIXjmcal/post # ln -s /disk1/FAIXJOBSC /opt/FAIXJOBSC/db # ln -s /disk1/FAIXMJS/var/spool/mjes /var/spool/mjes # ln -s /disk1/FAIXMJS/etc/mjes /etc/mjes # ln -s /disk1/FJSVstem /var/opt/FJSVstem (*1)
This is required only if the Master Schedule Management function is enabled.
Unmount the shared disk and deactivate the volume by using the varyoffvg command for AIX on the standby node.
# umount /disk1 # varyoffvg datavg1
